22
Soil Carbon Inventory and Working Lands Baseline Joshua Silva, Dr. Susan Crow, Dr. Johanie Rivera-Zayas, Elaine Vizka, Christine Tallamy Glazer University of Hawaii College of Tropical Agriculture and Human Resources

Soil Carbon Inventory and Working Lands Baseline

  • Upload
    others

  • View
    2

  • Download
    0

Embed Size (px)

Citation preview

Soil Carbon Inventory and Working Lands Baseline

Joshua Silva, Dr. Susan Crow, Dr. Johanie Rivera-Zayas, Elaine Vizka, Christine Tallamy GlazerUniversity of Hawaii College of Tropical Agriculture and Human Resources

Purpose

1. Compilation of available geospatially explicit datasets for soil carbon inventory and other attributes across natural and working lands (e.g., pasture, agriculture, agroforests)

2. Collect soil health, yield, and GHG sequestration data from a network of productive lands, farms, and ongoing trials as required for initialization of soil carbon-related planning tools

Scope of Work

Development of data resources required to generate a baseline and short/long term benchmarks for increasing GHG sequestration, soil health, and yields in natural and working lands (e.g., pasture, agriculture, agroforests) in Hawaii’s AFOLU sector

A. Thorough literature review

• GHG emissions, soil C storage

• Land use

• Soil C data

B. Compile geospatial datasets

• NCSS, NRCS, published/unpublished datasets

• SoilGrids (system for global digital soil mapping; machine learning)

• Land use datasets

Project Approach#1: Datasets

1A: Known References for GHG data in Hawaii’s NWLNatural, Working Land Sector

Management systems Management or Land cover References

Cropland Crops Intensive sugarcane cultivation

Conventional tillage and fertilizer management

Matson et al., 1996; Tran and Yanagida, 2019; Zachariassen et al., 1996; Pawlowski et al., 2017 and 2018

Tropical perennial grasses

Zero tillage, sugarcane and related bioenergy feedstocks, e.g., energycane, napier grass (Cenchrus purpureum),Guinea grass (Megathyrsus maximus) and others.

Pawlowski et al., 2017 and 2018; Meulemans, 2016; Crow et al., unpublished; Sumiyoshi et al., 2016

Biochar Biochar Meulemans, 2016; Biegert, 2015Organic Organic amendments Meulemans, 2016; Biegert, 2015Residue management Burning crop residues (what crop?) Miller et al., 1997

Aquaponics Vegetable production Wongkiew et al., 2018Fish production Hue et al., 2013

Forests Tropical rainforest Hall and Asner, 2007Montane forest Hedin et al., 2003 Fertility practices Hall and Matson, 1999 Invasive species Litton et al., 2006; Litton et al., 2008; Litton et al.,

2011; Hall and Asner, 2007Litter mineralization, and abiotic factors Riley and Vitousek, 2000; Holtgrieve et al., 2006

Forest fires Howbaker et al., 2017Peatlands/wetlands

Chimner, 2004; Beilman et al. unpublished

N=24

1A: Known References for GHG data in Hawaii’s NWL

Key Findings

• Croplands: Potential for napiergrass (reduced GHG emissions, water use); Biochar effect is soil dependent (Mollisol=reduced GHG emissions; Oxisol=increased GHG emissions)

• Aquaponics: High potential for N recovery from effluent via vegetables; Higher feeding rates correlated with increased N emissions

• Forests: N fluxes from nitrification in mesic tropical forests but denitrification in wetter sites; Soil organic C turnover about 0.39 Mg C/ha/year

• Pastures: GHG flux not directly quantified in Hawaii yet

1A: References for Soil C data in Hawaii’s NWLLand sector Management systems Land cover ReferencesForests Silvopasture Blackmore and Vitousek, 2000; Krueger and Ryals (unpublished)

Forest (non- specified, or diverse species)

Ares and Fownes, 2001; Burke et al., 2003; Scowcroft et al., 2004; McGrath 2019; Melone et al., 2021

Humid tropical forest Giardina et al., 2004Tropical Dry forest Elmore and Asner, 2006; Litton et al., 2006; Litton et al., 2008; Litton et al., 2011;

Chadwick et al., 2007Tropical rainforest Hall and Matson 2003; Hall and Asner, 2007; Rillig et al., 2001Montane forest Bothwell et al., 2014; Funk, 2005; Gower and Vitousek, 1989; Herbert and Fownes, 1999;

Hobbie, 2000; Idol et al., 2007; Kitayama et al., 1997; Riley and Vitousek, 1995; Rillig et al., 2001; Schuur et al, 2001; Selmants et al., 2014; Selmants et al., 2016; Chadwick et al., 2007; Hedin et al., 2003; Giardina et al., 2014

Native forest/ Forest reserves Austin, 2002; Austin and Vitousek, 1998; Chadwick et al., 2007; Chorover et al., 2004; Crews et al., 1995; Cusack et al., 2012; Hughes and Denslow, 2005; Hughes and Uowolo, 2006; Kao-Kniffin and Balser, 2008; Kramer et al., 2012; Mascaro et al., 2012; Neff et al., 2000; Osher et al., 2003; Sanderman and Kramer, 2013; Stewart et al., 2011; Giardina et al. 2014

Eucalyptus and Albizia Binkley et al., 1992; Kaye et al., 2000; Resh et al., 2002Eucalyptus plantation Giardina and Ryan, 2002; Binkley et al., 1992; Kaye et al., 2000; Resh et al., 2002; Ryan

et al., 2008; Crow et al., 2016; Zou and Bashkin, 1998Koa forest Idol et al., 2007; Litton et al., 2011; Scowcroft et al., 2004; Ares and Fownes, 2001;

Oʻhia forest (Metrosideros polymorpha)

Grant et al., 2019; Hobbie and Vitousek, 2000; Hughes and Uowolo, 2006; Kao-Kniffin and Balser, 2008; Kramer et al., 2012; Mascaro et al., 2012; Neff et al., 2000; Nusslein and Tiedje, 1999; Rilling et al., 2001; Sandermand and Kramer, 2013; Torn et al., 1997; Torn et al., 2005; Townsend et al., 1995; Townsend et al., 1997; Giardina et al., 2014

Fern (Dicranopteris linearis) Stewart et al., 2011Ōlapa (Cheirodendron trigynum) Stewart et al., 2011

Forest with Invasive species Litton et al., 2006; Litton et al., 2008; Litton et al., 2011; Melone et al. 2021Soil fertility/ Nutrient management practices

Giardina et al., 2003; Giardina et al., 2004; Gower and Vitousek, 1989; Hobbie , 2000; Hobbie and Vitousek, 2000; Neff et al., 2000; Ryan et al., 2008; Idol et al., 2007

Converted lands Abandoned to forest Pasture-abandoned/grassland-koa forest

Scowcroft et al., 2004; Idol et al., 2007

Plantation to Pasture, Secondary forest or forest

Bashkin and Binkley, 1998; Binkley and Resh, 1999; Binkley et al., 2004; Guo and Gifford, 2002; Kaye et al., 2000; Zou and Bashkin, 1998

Forest to Pasture, crop or managed forest

Guo and Gifford, 2002; Nüsslein and Tiedje, 1999

Pasture to Forest, secondary Forest, plantation or crop

Crow et al., 2016; Guo and Gifford, 2002

Intensive cultivation to perennial grass with zero tillage

Crow et al. 2020

1A: References for Soil C data in Hawaii’s NWLLand sector Management systems Land cover ReferencesAgricultural land Crop production Cusack et al., 2013

Sugarcane (Saccharum officinarum)

Burke et al., 2003; Pawlowski et al., 2018; Tirado-Corbalá et al., 2015

Napiergrass (Pennisetum purpureum)

Pawlowski et al., 2017 and 2018; Sumiyoshi et al., 2017

Guinea grass (Megathyrsus maximus)

Sumiyoshi et al., 2017

Energycane (Saccharum. officinarum x S. robustum cv. MOL-6081)

Crow et al. 2020

Orchards Coffee Youkhana and Idol, 2009; Youkhana and Idol, 2016Ash soils- no vegetation

Perez, 2001

Pasture Mixed Burke et al., 2003; Cusack et al., 2013; Chadwick et al., 2007Kikuyu pasture (Pennisetum clandestinum)

Cusack et al., 2012; Nusslein and Tiedje, 1999; Torn et al., 1997; Townsend et al., 1995; Townsend et al., 1997; Blackmore and Vitousek, 2000; Crow et al. 2016

Bufflegrass (Cenchrus ciliaris)

Torn et al., 1997

Grasslands Kramer and Chadwick, 2016; Scowcroft et al., 2004; Chadwick et al., 2007

Shrublands Kramer and Chadwick, 2016; Chadwick et al., 2007Shrubland Chadwick et al., 2007Peatlands/Wetlands

Beilman et al.

Hawaiʻi inventories and reports

Drawdown report 2020

1A: References for Soil C data in Hawaii’s NWL

Key Findings

• Croplands: Sugarcane plantations were a significant C source so potential for energy/fuel crops; Mulching and litter practices can increase soil C by 2.9 C Mg/ha

• Pasture: Importance of organic residues and minimized disturbance; Soil C stocks nearly similar between pasture and forest (9.5 kg C/m2 vs 12.7 kg C/m2), both more than sugarcane plantation

• No vegetation: dead tissue increased soil organic C vs living plants (Perez 2001)

• Converted Lands: Four years of perennial grass with zero tillage increased soil C stock from 18.0 kg C/ha to 22.6 kg C/ha in top 1m; Conversion of dry forest to non-native grass invasion reduced soil C storage at landscape scale

Hawaii Soil C data possibilities…

Courtesy E. Vizka

• Compared Soil Survey vs Additional Data

• 250m resolution

• 0-30cm depth vs 0-15cm depth

• Quantile Random Forest Model (5% and 95% quantiles)

• SoilGrids v1 (interpolate calculate)SoilGrids v2 (calculate interpolate)

*Limitations

o SoilGrids: under-estimate C for high organic soils

1B: Soil C Mapping

Summary: More data = More detail1B: Soil C Mapping Example

1B: Compiled Land Use DatasetsLand cover data layer Description CitationImportant Agricultural Lands (IAL) Classification based on importance of agricultural lands; integrates ALISH; criteria:

https://www.capitol.hawaii.gov/hrscurrent/Vol04_Ch0201-0257/HRS0205/HRS_0205-0044.htm

State Land Use Commission 2019

Carbon Assessment of Hawaiʻi Land Cover (CAH)

Land cover by biomes & invasion status; integration of HI-GAP, C-CAP, LF, and updates using very high resolution imagery

U.S. Geological Survey 2017

Agricultural Land Use Baseline (ALUB) Agricultural land use based on WorldView-2 satellite imagery (2011-2013), data provided by landowners and stakeholders, County Real Property Tax and Agricultural Water Use data; verified by site visits and stakeholder meetings.

Spatial Data Analysis and Visualization Lab 2015

Pre-contact Native Hawaiian Footprint Map of pre-contact Native Hawaiian land use based on archaeological evidence, information on native habitats, and natural condition information.

The Nature Conservatory & Office Hawaiian Affairs 2014

Coast Change Analysis Program Land Cover (C-CAP)

Land cover classification using multispectral analyses based on Landsat and high-resolution imagery; specifically for coastal lands

NOAA 1992-2012

LANDFIRE Vegetation (LF) Vegetation cover created by regression tree landscape models based on field data, satellite imagery, biophysical gradients

U.S. Geological Survey 2009

Gap Analysis Program Land Cover (HI-GAP)

Land cover using classification and regression trees based on Landsat TM satellite imagery 1999-2001, supplemented with Multi-Resolution Land Characteristic imagery and environmental data

Gon et al. 2006

Agricultural Land Use Maps (ALUM) Hand drafted maps from State Planning and Development Section & US Soil Conservation Service information; digitized

State Department of Agriculture 1978-1980

Agricultural Lands Importance (ALISH) Classified important agricultural lands into prime, unique, and other important lands; hand drafted; digitized

State Department of Agriculture 1977

Land Use Cover (LULC) Manual interpretation based on 1970's aerial photography U.S. Geological Survey 1976

Land Study Bureau (LSB) Land classification and productivity rating based on aerial photography and topographic maps; hand drafted onto paper; digitized

Land Study Bureau 1972

1B: Compiled Land Use DatasetsALUB vs CAH

• ALUB: 2011-2015, stakeholder input, only agricultural lands

• CAH: includes other table layers, 2014, NWL

ArcGIS 10.4.1

• “Joined” CAH & ALUB layers

• NAD 83 Zone 4N

Hawai`i Soil Health Data

• Ground-truth data points

• Collected 2017-2020 (up-to-date)

HSH land cover categories

ALUB categories

CAH land cover categories (major)

Organic cropland Diversified crop Agriculture

Conventional cropland

Seed production Grassland

Pasture Sugar/pineapple Shrubland

Unmanaged grassland

Flowers/foliage Forest

Agroforest Orchard OtherProtected forest Dairy Not vegetated

Unmanaged forest Pasture Developed

Commercial forestry

Wetland

Wetland taro

Aquaculture

HSH land cover categories

ALUB categories

CAH land cover categories (major)

Organic cropland Diversified crop Agriculture

Conventional cropland

Seed production Grassland

Pasture Sugar/pineapple Shrubland

Unmanaged grassland

Flowers/foliage Forest

Agroforest Orchard OtherProtected forest Dairy Not vegetated

Unmanaged forest Pasture Developed

Commercial forestry

Wetland

Wetland taro

Aquaculture

1B: Compiled Land Use Datasets

CAH: Larger general ag lands, but most pasture lands not classified ag

ALUB: Larger range of pasture lands, but did not distinguish developed areas (e.g.roads, structures)

HSH: Showed some ALUB lands were abandoned, unmanaged (up-to-date)

1B: Compiled Land Use Datasets

A. Collect soil from diverse natural and working land types

• Partnered projects, baseline data network

• Land use history and change; soil-carbon-building practices

Project Approach#2: Soil Health Data

B. Quantify soil health parameters

• Physical, chemical, biological (11 key parameters)

Samples Collected in Network

• # of sites= 14# samples= 47

• Land-use change (new) v. Benchmarks (long-term e.g.forested)

• Soil Types= Vertisols, Mollisols, Oxisols, Inceptisols

Compost, Organic Soil Amendments

Implemented Soil C Promoting Practices

Cover Cropping

Photo: Kako`o `Oiwi

AgroecologyRestoration

AnalysesRecommended Indicator Soil Function(s) Relation

Biological 24 hr CO2 burst Soil life, nutrient cycling, carbon storage and cyclingBeta-glucosiminidase

Beta-glucosidase

Mineralizable nitrogen

Chemical Total organic C % Carbon storage, soil life, nutrient cyclingDOC: DON

Hot water extractable organic C

pH

Physical Water holding capacity % Plant growth, soil life, water infiltration and supply, carbon cycling and storage

Water stable meta-aggregates %

Bulk density

Summarized Soil Health Results (Averages)Site PIAL. Current LU Min %OC CO2

burstPMN pH DOC:DO

NHWEC

WHC mega-WSA

BD

1 PIAL Unmanaged LAC 2.67 73.60 24.78 5.29 116.70 91.0 77.08 7.89 1.102 none Protect Forest LAC 6.68 371.26 175.92 7.22 13.27 1,223. 130.24 13.41 1.053 none Organic HAC 1.20 63.40 8.45 7.77 7.80 279.2 100.21 1.71 0.854 PIAL Convent. HAC 1.54 45.73 6.08 7.01 19.66 321.3 89.56 1.50 1.005 PIAL Convent. HAC 2.23 34.33 4.73 6.48 11.98 241.1 79.94 1.44 0.906 PIAL Convent. HAC 1.42 15.75 0.77 8.14 17.69 117.0 77.19 2.50 1.107 none Organic HAC 1.22 18.98 2.48 7.42 11.26 146.3 75.66 6.14 0.808 none Protect Forest LAC 15.16 133.16 78.79 5.15 88.84 1,172 129.50 39.66 1.109 PIAL Unmanaged LAC 10.04 405.27 124.28 7.64 159.19 1,286 104.94 10.16 1.10

10 PIAL Unmanaged LAC 7.86 371.20 111.55 6.91 113.06 454.2 107.40 7.95 1.1011 PIAL Unmanaged HAC 6.09 307.59 104.17 7.88 158.94 600.2 84.69 14.21 1.2512 PIAL Unmanaged HAC 10.39 69.95 29.12 5.89 47.34 155.3 133.61 11.38 1.2513 none Flood plain HAC 6.62 309.69 95.37 6.97 101.41 350.1 117.54 21.01 1.1314 N Beach Sand 25.88 49.04 21.87 7.91 135.43 147.5 42.00 0.59 1.48Ex PIAL --- --- 2.1 37.2 11.6 6.42 203.0 197.3 69.7 29.5 0.94Ex N --- --- 11.0 195.2 83.7 6.43 94.1 2,378.1 108.5 67.4 0.69

Source: Hubanks 2019

Summary• Gaps:

o GHG fluxes from pasturelands

o Soil C sequestration rates from ag lands and converted lands

o Inconsistencies in GIS data layers

• Future work (outside of this project)

o Updating NWL data layers to create tool

o Publications to assist land managers, soil C building practices

Mahalo!

Joshua Silva

UH Cooperative Extension

[email protected]

Dr. Susan Crow

UH CTAHR

[email protected]